What Occupancy Analytics Actually Measures
Plain description of what the sensors report, what has to be inferred from it, and where the inference usually goes wrong.
Basics · Explainer
Occupancy analytics counts presence in space over time. Everything else — utilisation, demand, whether a floor is worth keeping — is inferred, and the inference is where the decisions actually get made.
The space evidence in “What Occupancy Analytics Actually Measures” cannot explain by itself how project work is distributed or why a team uses the building differently. Used for how to measure employee productivity, how to measure employee productivity can add time and project context to aggregated occupancy findings, provided the two datasets keep separate purposes and are not merged into a hidden individual attendance score.
What is directly observed
Something was detected in a zone at a time. A body, a device, a badge swipe, a heat signature.
For a public, independent reference related to “What Occupancy Analytics Actually Measures”, consult the GSA workplace innovation resources. Its principles provide a useful check on scope, terminology, governance and the claims made during procurement or review.
For how long.
Sometimes how many, depending on the method.
That is the whole of the raw signal. Everything presented in a dashboard above that line involves assumptions.
What is inferred
That the detection was a person. Infrared sees a moving warm object; a cleaning trolley qualifies.
That one detection is one person. Device counting sees phones and laptops, and some people carry three.
That the space was in use. A desk with a bag on it registers as occupied by some methods and empty by others.
That absence means not needed. This is the largest inferential leap and it drives most of the expensive decisions.
The three numbers people confuse
Occupancy: how many people are present.
Utilisation: how much of the available capacity was used, over what period.
Capacity: how many could be there.
A room at four people out of twelve is 33% utilised and fully occupied for the purpose of booking, and both statements are true. Its own note covers the distinction, which is responsible for a surprising share of bad conclusions.
What this is for
Space planning: how much do we need, of what kind.
Operations: cleaning, catering, heating, lighting.
Design: which settings people actually use.
Not: who was where, how long individuals stayed, or whether somebody came in on Tuesday. That is a different activity with different obligations and it has its own note.
The honest framing
The data narrows uncertainty about how space is used.
It does not tell you whether the space is good, whether people would come in more if it were better, or what they would do if it were gone.
Programmes that treat the readings as answers rather than as evidence produce confident decisions on thin ground, which is the recurring failure in this field.
What this collection covers
What each counting method can and cannot see.
How to deploy sensors so the numbers mean something.
How to read the output without the common misinterpretations.
And where the line sits between measuring space and monitoring people, which is both an ethical question and a legal one.
What to check
Can you say what your sensors physically detect, as opposed to what the dashboard reports?
Do you know which of your numbers are measured and which are inferred?
Has anybody checked a reading against reality by standing in the room?
And is the stated purpose of your programme written down anywhere?